I am trying to run an older X11 server on a current Debian system. I have set up a /var/chroot/x11 with an older Debian with the desired version Xorg 7.1 (actually I want 7.2 but there's no Debian release with it). NVidia drivers are installed both in that tree and the root tree. I tried 195.36.24 and 256.35, all amd64.
That works fine, both when:
- running in the chroot using the chroot command (/dev /proc etc are all bind mounted)
- and also when calling the absolute path and using a different modulepath
X11 comes up, works, and the bugs that nagged me are gone. Yeah!
But libglx.so is not being loaded and no specific reason is given. It just says it can't load it. I am sure this isn't a chroot or absolute path related problem because a) the libglx.so.* is binary identical in both locations and b) all other modules, including nvidia_drv.so are loaded just fine.
Debugging this is difficult. Apparently my Linux here has decided that atime off is now the default and `ls -ltru` doesn't tell me what is being read and what is not. Nice. I tried `strace -f` but it doesn't follow the chain even though I'm root and not when using -u either
. Not sure what's up with that.
Here is the relevant passage from the logfile:
(II) LoadModule: "nvidia"
(II) Loading /var/chroot/x11/usr/lib/xorg/modules/drivers/nvidia_drv.so
(II) Module nvidia: vendor="NVIDIA Corporation"
compiled for 4.0.2, module version = 1.0.0
Module class: X.Org Video Driver
(II) LoadModule: "kbd"
(II) Loading /var/chroot/x11/usr/lib/xorg/modules/input/kbd_drv.so
(II) Module kbd: vendor="X.Org Foundation"
compiled for 7.1.1, module version = 1.1.0
Module class: X.Org XInput Driver
ABI class: X.Org XInput driver, version 0.6
(**) Jun 28 16:52:05 NVIDIA(0): Enabling RENDER acceleration
(**) Jun 28 16:52:05 NVIDIA(0): Using static 90-degree clockwise screen rotation.
(EE) Jun 28 16:52:05 NVIDIA(0): Failed to initialize the GLX module; please check in your X
(EE) Jun 28 16:52:05 NVIDIA(0): log file that the GLX module has been loaded in your X
(EE) Jun 28 16:52:05 NVIDIA(0): server, and that the module is the NVIDIA GLX module. If
(EE) Jun 28 16:52:05 NVIDIA(0): you continue to encounter problems, Please try
(EE) Jun 28 16:52:05 NVIDIA(0): reinstalling the NVIDIA driver.
(II) Jun 28 16:52:06 NVIDIA(0): NVIDIA GPU GeForce GTX 275 (GT200) at PCI:4:0:0 (GPU-0)
So in conclusion to me this looks like the libglx.so.* (this comes binary in the package, right?) is being refused by the Xorg server, for example due to missing symbols, but the error messages are hidden from me.
Any suggestions on how to proceed?
log'n'config appended and also as plain text on: